Distributed Database Commit Timestamping for Ordered Binary Logs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed databases, operations on different storage nodes can lead to incorrect ordering of transactions, resulting in inconsistent data due to the independent generation of binary logs at each node, which lack a chronological order across nodes.
Innovation Solution
A distributed database system with a computing module that sends commit instructions with globally ordered timestamps to storage modules, ensuring that binary logs are generated with a consistent chronological order, thereby maintaining data consistency across the system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If each storage node independently generates binary logs, then each node can operate autonomously and process transactions locally, but the binary logs from different nodes lack chronological order across the distributed system
Solution Approach 1:
The patent introduces a distributed transaction coordinator as an intermediary component that manages transaction execution across multiple storage nodes. The coordinator assigns unique transaction IDs and timestamps to each transaction, ensuring that binary logs from different nodes can be chronologically ordered by referencing these centralized identifiers. This mediator resolves the conflict between autonomous node operation and global ordering requirements.
Solution Approach 2:
The patent modifies the binary log structure by adding global transaction IDs and timestamps as new parameters. Each binary log entry now includes references to the coordinator-assigned transaction identifier and timing information, transforming the previously node-local logs into globally orderable records. This parameter enhancement enables chronological sorting across distributed nodes without changing the fundamental autonomous operation model.
2Adaptability or versatility
If binary logs from different storage nodes are directly subscribed to, then data can be propagated across the distributed system, but incorrect ordering of operations occurs leading to data inconsistency
Solution Approach 1:
The patent implements a feedback mechanism where the distributed transaction coordinator monitors transaction execution status across all storage nodes and adjusts the ordering of binary log propagation accordingly. By continuously gathering information about transaction completion states and using this feedback to determine propagation timing, the system ensures that data is propagated in the correct chronological order, maintaining consistency while enabling versatile data distribution.
Solution Approach 2:
The patent applies preliminary action by having the distributed transaction coordinator pre-assign transaction IDs and establish execution orders before transactions are actually executed at different storage nodes. This advance planning ensures that when binary logs are generated and propagated, they inherently carry the correct ordering information, preventing consistency issues before they can occur during data propagation.
3Productivity
If operations on different storage nodes are allowed to proceed independently, then system throughput is improved, but the lack of global ordering causes downstream applications to receive incorrectly ordered data
Solution Approach 1:
The patent adds a new dimension to binary log ordering by introducing global transaction IDs and timestamps that operate perpendicular to the local node operation timeline. Instead of relying solely on local sequential numbering, each binary log entry now carries coordinates in a global temporal dimension, enabling downstream applications to correctly order operations from multiple nodes simultaneously while maintaining high throughput independent execution.
Data Source
AI summary
A distributed database system includes: a computing module configured to receive a transaction starting request; determine a target transaction and a target storage module corresponding to the target transaction based on the transaction starting request; send an execution instruction to a target storage module; send a commit instruction to the target storage module when the target storage module returns a message that the target transaction is executed successfully, the commit instruction including a commit timestamp of the target transaction; and the target storage module configured to receive the execution instruction, and execute the target transaction according to the execution instruction; return the message that the target transaction is successfully executed to the computing module when the target transaction is successfully executed; and receive the commit instruction, commit the target transaction according to the commit instruction, and generate a binary log according to the commit timestamp of the target transaction.